Eye Tracking Case Study Answers

1167 Words3 Pages

When the participant arrived, the rules and the route that they had to drive were explained and they were also asked to sign a consent document. The rules were that they should never take their hands off of the steering wheel. They should always keep both their hands at the same height unless they had to turn the wheel. Each participant had to do four trials. The first trial was more of a warmup trial so that the participant gets to understand when and where they had to change lanes. For the first trial, none of the CAT devices were turned on. However, for trials 2, 3 and 4, one of the CAT devices were activated. Participants one to five were tested with Garmin, Participants six to ten with Audiovox, Participants eleven to sixteen with Mobileye and the final five participants with RD-140. …show more content…

Data Analysis The different types of warnings the CAT device gave were noted, according to the time of a trial. For example, if the CAT device gave out a Forward Collision Warning at the fifth minute of the trial, then it was noted as FCW as a warning, and the time it was shown as 05:00:00. This was used later on to find the exact time and place the CAT device gave the warning by using eye tracking data. By using eye-tracking data, it was used to figure out whether the warning that the CAT device gave was correct or not. The warning was judged to be correct whenever the driver took actions by either slowing down by braking or steering the wheel. For example, if the CAT device gave a Forward Collision Warning and the driver took action by slowing down, then it was judged to be a correct warning. However, if there was no vehicle in front and the driver chose to ignore it, then it was judged to be a false warning. Figure 3.5 shows that the CAT device, RD-140 gave a Forward Collision Warning. The eye-tracking data revealed that there was a vehicle in front and the participant took action by slowing down in the next few seconds. Therefore, this was calculated as a correct

Open Document